Introduction:

Roselle (Hibiscus sabdariffa) is considered an economically (^{Chew et al., 2024}) and medicinally (^{Takada et al., 2024}) important species. On an international scale, Mexico ranks seventh as a producer, with 5.14%. The state of Guerrero is the largest producer of roselle in the country. The municipal areas of Ayutla de los Libres and Tecoanapa are the main production areas, followed by Acapulco de Juárez and San Luis Acatlán (SIAP-SADER, 2022). However, the sale price is low, due to the quality and innocuousness since preharvest, due to problems caused mainly by phytopathogenic fungi (^{Aparicio et al., 2016}; ^{Dios-López et al., 2011}).

Among the most important diseases affecting roselle in Guerrero are the calyx spot induced by Corynespora cassiicola (^{Ortega-Acosta et al., 2015}a), Phytophthora parasitica (^{Hernández-Morales and Romero-Cova, 1990}), Lasiodiplodia pseudotheobromae, Fusarium incarnatum-equiseti (^{Aparicio et al., 2016}) and other phytopathogenic fungi associated to these symptoms, such as F. oxysporum, F. solani and F. incarnatum (Ortega-Acosta et al., 2015b). Recent studies indicate that the roselle spot disease is the main constraint on calyx production, since it displays a 100% incidence in Ayutla and Tecoanapa (Ortega-Acosta et al., 2020).

Studies performed on calyces report that out of 1,470 kg of roselle gathered in Guerrero, 40.8% were of commercial quality, with spots and smut on the tips (equivalent to 588 g of the unit) with calyxes under bad innocuousness conditions. The phytopathogenic fungi related to roselle in storage were Fusarium and Corynespora (^{Ruiz-Ramírez et al., 2015}).

Mexico has a wide variety of temperate climate coniferous and broadleaf tree species (^{Quiroz y Magaña 2015}), with a potential in the use of organic extracts with antifungal effects against phytopathogenic fungi, able to reduce diseases, with favorable effects on the quality of the calyces, including pine nut and linaloe, Jatropha curcas and Bursera linanoe respectively; they are considered a new organic alternative to reduce the use of chemical fungicides (^{Del Puerto et al., 2014}). The aim of this investigation was to evaluate the inhibitory effect of J. curcas and B. linanoe resins, the mixture of both, and thiabendazole against C. cassiicola, F. oxysporum and F. solani under in vitro conditions using the Kirby Bauer and agar well diffusion techniques.

The J. curcas resin sample was gathered fresh and the linaloe (B. linanoe) resin was taken from the collection of the National Forestry, Agriculture, and Livestock Research Institute (INIFAP), Zacatepec, Morelos experimental field. Both resin samples were taken from the fruit, which were placed in sterile jars with a screw-on lid for their transportation under refrigeration at a 20 °C.

The C. cassiicola with Access in GenBank (MF000865) responsible for roselle spot, F. oxysporum (KM519188) and F. solani (KM519190) associated to the “black leg” symptom, originate from the roselle cultivation in the state of Guerrero, and were provided by the Laboratory of Plant Health and Safety of the Phytosanitary Institute at the Colegio de Postgraduados-Campus Montecillo.

To determine the in vitro effect of the resins on the fungi, the strains were planted in Petri dishes in a Potato-Dextrose-Agar (PDA) medium at room temperature (25 ± 2 °C). After 7 days, it was covered with 10 mL of sterile saline solution at 0.85%, shaken for 3 minutes and rubbed on the surface of the culture, firmly and gently, using a bacteriological loop. The suspension obtained from each one of the strains was poured directly into sterile test tubes and left to rest for 20-30 min. The turbidity of the spore suspension was measured by spectrophotometry, 70 75% transmittance at 530 nm, according to the technique by ^{Cermeño and Torres (1998}).

For the agar well technique, the technique followed was the one proposed by ^{Moreno-Limón et al. (2011}). For the Kirby-Bauer technique, the method by ^{Gallardo-Vásquez et al. (2019}) was followed, using Petri dishes with a PDA culture medium. Separately 10, 20 and 30 μL of the de la spore suspension of each of the fungi was planted, at a concentration of 1x10¹ with four repetitions.

The Petri dishes were incubated at 25 ± 2 °C for 7 days. Subsequently, the diameter of the radial growth was calculated. The percentage of growth inhibition was calculated considering that the inhibition is the opposite of the growth (TequidaMeneses et al., 2002). The inhibiting effect of the resins was determined to be positive with the appearance of a zone of inhibition around the perforations, which was measured using a 150 x 0.02 mm Fumetax^® caliper every 24 h for 7 days.

The minimum inhibitory dose (MID) was determined, which is defined as the lowest concentration of resins that showed no growth in the test tubes. PDA culture medium was used, in which 8 tubes were inoculated with 1 mL of PDA with 10 μL of spore suspension, at a concentration of 1x10¹, adjusted at 530 nm, 70 75% of transmittance (^{Cermeño and Torres 1998}).

Statistical analysis. The sample averages of the studied factors were calculated and added to their standard deviations in parentheses. Box graphs were also created to know their distribution. The experiment followed a completely randomized factorial design with fixed effects. Four factors were studied: technique (Ti) with two levels: agar disk (Kirby Bauer) and agar well diffusion; resins (Rj) with four levels: B. linanoe, J. curcas, the mixture of both, and thiabendazole, volume (Vk) with three levels: 10, 20 and 30 µL and fungi (Hl) with three species: C. cassiicola, F. oxysporum and F. solani, and the response variable was the diameter of the inhibition halo. The statistical model used was:

Yijklt = μ + Ti + Rj + Vk + Hl + (TR)ij + (TV)ik + (TH)il + (RV)jk + (RV)jl + (RH)kl+ (TRV)ijk + (TRH)ijl + (RVH)ikl + (RVH)jkl + (TRVH)ijkl + εijklt

ε ~ IIDN(0, σ ²); i = 1, 2; j = 1, 2, 3; k = 1, 2, 3; l = 1,2, 3, 4; t = 1, 2, 3.

The data were analyzed using the R statistical package R (R Core Team, 2020). In the analysis of variance, a significance level of α = 0.05 was used. The nonparametric aligned ranks version was used to determine whether the main effects and interactions were significant (^{Kay et al., 2021}). To compare the 72 treatments, the non-parametric Kruskal-Wallis test was used.

A modification was performed on the minimum inhibitory concentration (MIC) since, due to the handling of raw and natural samples without lab equipment, their concentrations are unknown; only different doses were handled (Table 1). The B. linanoe displayed a better MID in comparison with the rest of the samples (Table 1).

The 20 treatments show a predominance of the agar well technique, since it appears in 12 treatments; the B. linanoe resin treatment is amongst the 20 best treatments, since it appears in nine treatments; the mixture of resins in five treatments, and thiabendazole, in six treatments; the 10 and 20 µL volumes appear in eight and seven treatments respectively, and the fungi C. cassiicola and F. oxysporum appear in 12 and eight treatments, respectively. The combination of agar well and B. linanoe appeared in six of the 20 treatments.

On the opposite end, the group with the lowest means of the inhibition halo diameter contains the five treatments that had a minimum of the inhibition halo diameter of 13 mm (DJcV30Fo, DTV10Fs, DTV30Fs, PTV10Fs and PTV30Fs) and another 14 treatments (DJcV10Cc, DTV20Fo, DJcV30Cc, DTV30Fo, PTV30Fo, DJcV20Cc, DJcV20Fs, PJcV20Fo, DJcV30Fs, PJcV30Fo, DTV20Fs, PTV20Fs, DJcV10Fs, DJcV20Fo), making up a total of 19 treatments. In these treatments, the agar disk technique was predominant, since it appears in 13 treatments; the J. curcas resin appears in 10 treatments and thiabendazole, in nine treatments. The 30, 20 and 10 µL volumes appear in eight, seven and four treatments, and the microorganisms F. solani, F. oxysporum and C. cassiicola appear in nine, seven and three treatments, respectively (Table 2).

All factors were found to be significant, including the main effects of the technique, the resins, the volume and the fungi in the diameter of the inhibition halo, as well as the double, triple and quadruple interactions. The B. linanoe resin displayed the highest inhibition, and therefore the greatest antifungal effect against C. cassiicola and F. oxysporum. This antifungal effect can be related to a variety of secondary metabolites found in the plant, such as saponins, phenols, flavonoids and monoterpenic compounds such as the linalyl acetate (^{Becerra and Noge, 2010}; ^{Cruz et al., 2016}). The antifungal response obtained from B. linanoe is consistent with ^{Arellano-Hernández (2018}), who in recent studies observed that the resins obtained from different parts of the tree present bacteriostatic and bio-fungicidal properties.

This aligns with ^{Seepe et al. (2020}), who mentioned that the plant extracts have presented a high antifungal activity against different pathogens of the Fusarium genus. The metabolites of the resins, such as linalyl acetate, caryophyllene and undecane, reported in linalool (^{Cruz et al., 2016}).

The case of the J. curcas resin did not provide promising results, showing that with the agar well technique, an inhibition halo was obtained in the fungus C. cassicola, between 12.0 and 13.5 mm in diameter. Despite ^{Córdova-Albores et al. (2016}) mentioning that the effect of the Jatropha curcas oil has been widely studied, it has been found to have the ability to partially alter the morphology of the mycelium and conidia of F. oxysporum, causing morphological and cellular damage, including intense vacuolization. These effects are attributed to the large number of metabolites with antifungal activity found in diverse parts of the plant (^{Saetae and Suntornsuk, 2010}). ^{Rahu et al. (2021}) also mention that it is crucial that the trials determine the fungicidal effect of the extracts or compounds, since J. curcas is a plant with a great potential as an antimicrobial agent.

On the other hand, the mixture of both resins (B. linanoe + J. curcas) to boost the antifungal effect was lower than expected, presenting inhibition halos of 16.1 to 18.7 mm, greater than those found with the J. curcas resin, which presented inhibition halos of 7 to 13.1 mm, though lower than those with B. linanoe, with inhibition halos with a diameter of 30 mm. These are marginal means that do not consider the effect of the other three factors studied. Therefore, mixing both resins is not considered a good option. The best minimum inhibitory concentration was for B. linanoe (4 µL mL⁻¹), followed by the mixture (12 µL mL⁻¹), and finally, J. curcas (140 µL mL⁻¹) (Table 1).

The commercial control thiabendazole displayed control in six of the treatments, presenting an inhibition halo of 30 mm with the agar well technique of the fungi C. cassiicola and F. oxysporum. This coincides with ^{Medina-Osti et al. (2022}), who mention that thiabendazole has shown its antifungal action derived from its direct effect on mitosis, which hinders the development of the mycelium, therefore its efficiency in the reduction of the growth of the mycelium in species of Fusarium. However, the use of thiabendazole, one of the most efficient fungicides against a wide variety of pathogens, is no longer an effective treatment, although some farmers still use it to control some diseases caused by pathogens (^{Seepe et al., 2021}).

Regarding every fungal species, the largest inhibition halos, with 30 mm, were found in C. cassiicola, being one of the most susceptible and with the least resistance, since it displayed greater zones in all the volumes used. F. oxysporum also displayed susceptibility with a 30 mm inhibition halo, presenting larger zones in the fungal volumes of 10 and 20 µL. This suggests that the use of these resins may be a viable alternative against the roselle leaf and calyx spots. These findings coincide with those by ^{Moreno-Limón et al. (2011}), who evaluated different Aspergillus and Penicillium genera spore suspensions, observing a high variability in their analysis. However, the reports on C. cassiicola in roselle are scarce in the state of Guerrero (^{Ortega-Acosta et al., 2015}a). It is therefore important to continue researching to understand the specific action mechanisms of the Bursera linanoe resin.

No significant differences were found in the evaluations of the volumes of fungi used, since the largest inhibition halo diameters were consistently observed in the volumes of 10, 20 and 30 µL. The B. linanoe resin, obtained from the fruit of the tree, is considered a promising and efficient alternative as a biofungicide in the control of C. cassiicola, F. oxysporum and F. solani, according to the in vitro results. Therefore, investing in the development of medicinal plant-based products for the control of crop diseases caused by pathogens is a growing sector that should be considered and developed (^{Seepe et al., 2021}).

Reducing the use of conventional synthetic fungicides by incorporating natural effective products is a crucial step towards a sustainable agricultural production. Nevertheless, it is crucial to understand the action mechanism of these resins and evaluate their effectiveness under field conditions. It was found that all terms were significant, including the main effects of technique, resins, fungal volume and fungal species on the diameter of the inhibition halo, as well as the double, triple and quadruple interactions.

The results of this in vitro investigation show that the B. linanoe resin exhibited greater inhibition for C. cassiicola and F. oxysporum in both techniques (the agar well technique and agar disk or the Kirby Bauer technique), making it the type of resin with the greatest biocontrol potential.